Osteoarthritis (OA) is the most common joint disorder. Radiographic changes of OA occur in most people by 65 years of age and are present in more than 80% of people over age 85. OA is most common in the hands and is also common in the knees where symptoms are often severe. Studies looking at the causes of OA and the increase in OA with age have in general been cross-sectional. There have been few attempts to follow individuals over time to assess disease development. The objective of this project is to examine the long term course and risk factors for osteoarthritis. Specifically, we will look at the natural history of OA in a population based group of elderly, the Framingham Heart Study cohort. This group obtained hand X-rays 25 years ago and knee x-rays 8 years ago as part of the Framingham OA Study. The investigators propose to do a follow-up study. The specific aims of this project are to look at the natural history of disease and risk factors for OA. This study would involve interviews with subjects along with physical examinations and x-rays of the knees and hands. The analyses would focus on whether OA has developed or progressed since the first exam. For those with new or progressive knee OA or hand OA, the investigators propose to examine what risk factors might have caused disease. They will focus on age, sex, obesity, chondrocalcinosis, and osteopenia as identified by bone mass studies also being performed in the group. They would also investigate potential new risk factors including walking activity, knee alignment, and hormones (including gonadal hormones and insulin-like growth factor I). Also, they would look at whether there is a generalized disease process in subjects after controlling for known causes of disease and investigate whether subjects with hand or knee OA tend to develop OA in the other site more frequently. Because it is longitudinal, the investigators believe that the project offers potentially new and important insights into all aspects of OA epidemiology. They state that the study of natural history may identify subgroups in whom disease is likely to develop or progress. They also state that the study of risk factors will be uniquely valuable because it is longitudinal and the proposed inquiry into potential unstudied risk factors may yield major new insights into causes of disease.